genetic engineering and genomics - garland · pdf file · 2017-07-14genetic...

Genetic Engineering and G

enomics

Issues

Biological Concepts

Chapter Outline• Does genetic engineering

fundamentally change thebiology of an organism?

• Does gene therapy work?

• When should gene therapy be used? When should it not be used?

• Do DNA tests positively identify individuals?

• Why does the U.S. government fund the Human Genome Project?

• What benefits have been derived from the Human Genome Project?

• How could the results of the Human Genome Project be misused? How can we guard against such misuse?

• Biotechnology (The Human Genome Project; genetic engineering)

• Molecular biology (genomics; bioinformatics)

• Structure–function relationships (proteomics)

Genetic Engineering Changes theWay That Genes Are Transferred

Methods of genetic engineering

Genetically engineered insulin

Gene therapy

Molecular Techniques Have Led toNew Uses for Genetic Information

The first DNA marker: restriction-fragmentlength polymorphisms

Using DNA markers to identify individuals

Using DNA testing in historical controversies

The Human Genome Project HasChanged Biology

Sequencing the human genome

The human genome draft sequence

Mapping the human genome

Some ethical and legal issues

Genomics Is a New Field of BiologyDeveloped as a Result of the HumanGenome Project

Bioinformatics

Comparative genomics

Functional genomics

Proteomics

4

95

96

As a result of information published in 2001, humans now know moreabout themselves, at least at the molecular level, than they ever have

before. This watershed date marked the publication of the draft of thenucleotide sequence of all of the DNA in human chromosomes. Along theway, a complete map of the location of these nucleotide sequences on thechromosomes was also produced. All of this information is stored in anenormous database that is publicly available for use by any scientist inthe world. A tremendous amount of basic molecular biology has beendiscovered in the course of the Human Genome Project that producedthis database. As a tool for biological research, this database potentiallyoffers new ways of studying everything else in biology. In addition, theproject has spawned many practical advances in biotechnology andgenetic engineering.

Genetic Engineering Changes the Way ThatGenes Are Transferred

Genetic engineering is the direct alteration of individual genotypes. It isalso called recombinant DNA technology or gene splicing, terms whichare used interchangeably. Human genes can be inserted into human cellsfor therapeutic purposes (gene therapy, p. 100). In addition, because allspecies carry their genetic information in DNA and use the same geneticcode, genes can be moved from one species to another. The uses ofgenetic engineering in plants are discussed in Chapter 11. Here we seesome of the applications of genetic engineering for human medicine.

Methods of genetic engineeringWhether the ‘engineered’ gene is one from the same species or a differentspecies, the techniques are much the same. All these technologies dependon being able to cut and reassemble the genetic material in predictableways. This is possible owing to the discovery of special enzymes calledrestriction enzymes.

Restriction enzymes. Restriction enzymes are enzymes used to cutDNA at specific sites. There are several hundred restriction enzymes cur-rently known and each cuts DNA at a different nucleotide sequence;these target sites are generally about four to eight nucleotides long (Fig-ure 4.1). Each of these restriction enzymes is a normal product of a par-ticular bacterial species, and most are named after the bacteria fromwhich they are derived. Thus, in Figure 4.1, HaeIII is an enzyme from thebacteria Haemophilus aegypticus and EcoRI is from Escherichia coli.They are called restriction enzymes because their normal function withinthe bacteria is to restrict the uptake of DNA from another bacterialspecies. Each species’ restriction enzyme cuts the DNA from otherspecies, but not its own, because its own DNA does not contain thenucleotide sequence that is the target site for its own enzyme.

Genetic Engineering and Genomics4

Figure 4.1Restriction enzymes. Thenucleotide sequencesrecognized and cut by therestriction enzymes HaeIIIand EcoRI are shown.

Genetic Engineering Changes the Way That Genes Are Transferred 97

Several other enzymes are known that can break apart a DNAmolecule, but an enzyme that acts indiscriminately is of little use ingenetic engineering. Restriction enzymes act specifically. Each restric-tion enzyme generally cuts a sample of DNA in several places, whereverthe DNA contains a particular sequence of bases that the enzyme recog-nizes, forming a series of pieces (called restriction fragments). A givenrestriction enzyme mixed with the same sequence of DNA always pro-duces the same number of fragments. The length of the pieces may varyif there are variable repeat sequences, for example, but the number ofpieces and the places cut are always the same. Before the discovery ofrestriction enzymes, breaking chromosomal DNA into pieces was donemechanically, producing different numbers of pieces every time the pro-cedure was done, making the results of DNA techniques impossible toreproduce from one experiment to the next. Because restriction enzymesalways cut at the same sites, they can be used in genetic engineering.

Restriction enzymes in genetic engineering. The first step in insertinga gene for genetic engineering is to isolate the gene in question. This iscarried out by using a restriction enzyme to snip out the desired segmentof DNA. Each restriction enzyme cuts the DNA at specific places, definedby their DNA sequences. The most useful restriction enzymes are thosethat cut the two DNA strands at locations that are not directly across fromeach other, producing short sequences of single-stranded DNA known assticky ends (see Figure 4.1). For example, the commonly used restrictionenzyme EcoRI always targets the sequence GAATTC, cutting it between Gand AATTC, breaking the two-stranded sequence into fragments that havesticky ends. The ends are called ‘sticky’ because they can stick togetherspontaneously with another molecule containing complementary stickyends. In fragments cut with EcoRI, the single-stranded AATT sequencescan pair with one another, stick together, and then be joined permanently.(An enzyme such as HaeIII that cuts at sites directly across from eachother forms ‘blunt’, rather than sticky, ends, as shown in Figure 4.1.)

If a particular restriction enzyme produces sticky ends, all fragmentscut with that enzyme will have sticky ends that match one another. Thus,a fragment can be joined to any other fragment cut with the sameenzyme. This makes it possible to use restriction enzymes to cut a DNAsequence and insert a functional gene with matching sticky ends.

G

C G

C

G

C

GG

HaeIII

EcoRI+

+

C

C

T

A

G

C

T

TAA

T

A

target sitesugar–phosphate backbone

double-strandedDNA

The HaeIII target site is fourbases long and the enzyme cutsthe DNA strands at sites directlyacross from each other, leavingdouble-stranded ('blunt') ends.

The EcoRI target is six bases long and it cuts the DNA between G and AATTC. The sites on the two strands are not directly across from each other, leaving short single-stranded ('sticky') ends.

G

C

G

CG

C

G

C

G

C T T A A G

CA TA T

Chapter 4: Genetic Engineering and Genomics

Restriction enzymes that produce blunt ends are useful in other ways,but are not useful for genetic engineering because the fragments cannotbe put back together.

Cutting an entire chromosome with a restriction enzyme producesmany fragments, only one of which contains the gene to be isolated. ADNA probe specific for the gene will isolate the fragment containing thegene of interest. As we have seen before, such a probe is a complemen-tary DNA strand that carries a radioactive or chemical tag. The probeallows geneticists to isolate the labeled sequences, and then separate thedesired genes from the DNA probes that pair with them.

A functional gene isolated in this way can then be inserted intoanother piece of DNA. The target DNA is cut with the same restrictionenzyme, so sticky ends complementary to the fragment are available, andthe gene can be incorporated permanently. So far, most genetic engineer-ing of human genes has involved the introduction of these human genesinto bacteria. The reasons for this are largely practical: many humangene products are useful in medicine but are more readily produced inlarge amounts inside genetically engineered bacteria than inside people.For example, the hormone somatostatin, also called growth hormone, ishighly valued for the treatment of certain types of dwarfism. The hor-mone is, however, difficult to obtain from human sources (the traditionalway is to extract it from the pituitary glands of dozens of cadavers) and istherefore very expensive. Insulin, the hormone needed by diabetics, isanother example of a human gene product. Both of these hormonescould be obtained from sheep or pigs or other animals, but the animalhormones are not as active in humans as the human hormones, andsome patients are allergic to hormones obtained from other species.Genetic engineering provides a cost-effective way of manufacturing largeamounts of these human hormones in bacteria.

Genetically engineered insulinHuman insulin was the first commercially produced genetically engi-neered product. The initial step is to grow human cells in tissue culture.Tissue culture is a procedure in which cells that have been removed froman organism are grown in a dish of nutrient-rich medium kept at bodytemperature in an incubator. After a sufficient number of cells havegrown, DNA extracted from the cell nuclei is then exposed to a restrictionenzyme that cuts the DNA into desired fragments. One fragment containsthe human gene for insulin, which can be isolated using a DNA probe.

The same restriction enzyme is used on nonchromosomal DNAmolecules, called plasmids. Bacteria have a single chromosome in theform of a closed loop. Many also have a number of plasmids, short cir-cular DNA pieces that are separate from the bacterial chromosome(Figure 4.2). Plasmids are used in genetic engineering because, beingshort, they have fewer sites at which a given restriction enzyme can cut.Cutting a DNA sequence in the plasmid with the same restriction enzymethat was used on the human DNA creates sticky ends that match theDNA fragment taken from the human cell. This allows incorporation ofthe human gene for insulin into the bacterial plasmid. The bacteria arethen treated so that they take up the engineered plasmid. In most cases,the plasmid also contains another DNA sequence that can be used to

98

Figure 4.2A bacterial cell showing itssingle chromosome and oneplasmid.

bacterialchromosome plasmid


select the bacteria that have incorporated an engineered plasmid. Forexample, the plasmid might contain the gene for an enzyme that givesthe bacteria resistance to a common antibiotic; the antibiotic can then beused to select the bacteria that have incorporated this gene while killingthe majority that are still susceptible. The procedures sound easy andstraightforward, but each step of the process is technically difficult andonly a small proportion of the attempts succeed.

The genetically altered bacterium can now be cloned, that is, allowedto multiply asexually, which produces vast numbers of genetically identi-cal copies of itself and its engineered plasmid. The resultant bacteriathen transcribe and translate the human gene to produce human insulin(Figure 4.3). The human insulin extracted from these bacteria, calledrecombinant human insulin, can be given to diabetic patients.

Figure 4.3Production of geneticallyengineered insulin.

Isolate human cellsand grow in tissueculture.

1

Isolate DNA from the human cells.2

Use a restriction enzyme to cut DNAinto fragments with sticky ends. Isolatethe fragment containing 'insulin gene'with a probe.

3 Meanwhile, isolate plasmid DNA froma bacterium.

4

Use the same restriction enzyme to cut theplasmid DNA, creating matching sticky ends.

5

Combine plasmid and human DNA; someof the plasmids will recombine with thehuman DNA fragment containing theinsulin gene.

6

Allow new bacteria to incorporate the recombinant plasmid into the bacterial cell, then screen bacteria to find the ones that have incorporated the human gene for insulin.

7 Grow trillions of new insulin-producing bacteria.

8

etc.


Gene therapyInstead of growing human insulin in bacteria (see Figure 4.3), geneticengineering could theoretically be used to introduce the insulin gene intohuman cells that do not possess a functional copy. (That would still notcure diabetes unless these cells were also capable of appropriatelyincreasing or decreasing their output of insulin according to conditions.)This type of genetic engineering is called gene therapy, the introductionof genetically engineered cells into an individual for therapeutic purposes.

Treatment for hereditary immune deficiency. Human gene therapyhas been used successfully to treat severe combined immune deficiencysyndrome (SCIDS), a severe and usually fatal disease in which a child isborn without a functional immune system. Unable to fight infections,these children will die from the slightest minor childhood disease unlessthey are raised in total isolation: the ‘boy [or girl] in a bubble’ treatment.The enzyme that controls one form of SCIDS has been identified; it iscalled adenosine deaminase (ADA) and its gene is located on chromo-some 20. A rare homozygous recessive condition results in a deficiencyof this enzyme, which in turn causes the disease.

Gene therapy for this condition consists of the following proceduralsteps, shown in Figure 4.4.

1. Normal human cells are isolated. The cells most often used are Tlymphocytes, a type of blood cell that is easy to obtain from bloodand easy to grow in tissue culture.

2. The isolated cells are grown in tissue culture.

3. The DNA from these cells is isolated.

4. A restriction enzyme is used to cut the DNA into fragments withsticky ends; one will contain the functional gene for ADA. A probewith a complementary DNA sequence is then used to isolate andidentify the fragments bearing the gene.

5. The same restriction enzyme is used to create matching stickyends in viral DNA isolated from a virus known as LASN. This viruswas chosen because it can be used as a vector: it can transfer thegene into the desired human cells—the host. (Other vector viruseshave also been used; each virus type varies in the size of DNA frag-ment that can be inserted and the type of cell that it can enter.)

6. The viral DNA is then mixed with the human DNA fragments andallowed to combine with them.

7. The virus is allowed to reassemble itself; it is then ready for further use.

8. Blood is drawn from the patient to be treated and T lymphocytesare isolated from this blood. These lymphocytes, like all of theother cells from this person, are ADA-deficient because they donot possess a functional ADA allele.

9. The virus is now used as a vector to transfer the functional gene.The virus must get the gene not only into the lymphocyte but alsointo its nucleus. The gene must incorporate into the cell’s DNA ina location where it will be transcribed and where it does not breakup some other necessary gene sequence.

100


10. The lymphocytes are tested to see which ones are able to producea functional ADA enzyme, showing that they have successfullyincorporated the functional ADA allele.

11. The genetically engineered lymphocytes are injected into thepatient, where they are expected to outgrow the genetically defec-tive lymphocytes because the ADA-deficient cells do not divide asfast as cells with the ADA enzyme.

Figure 4.4An example of gene therapyshowing the transfer of thehuman gene responsible foradenosine deaminase (ADA).

Isolate normal humanT lymphocytes.

1 Grow lymphocytes in tissue culture.

2 Isolate DNA from someof the cells.

3

Use a restriction enzyme to cut this DNA and produce 'sticky ends', then isolate fragment containing the gene for ADA enzyme.

4

Also isolate DNA from the LASN virus and cut with the same restriction enzyme.

5

Mix the DNA fragments.6

Combine T lymphocytes withLASN virus (vector).

9

Grow cells and test for the ADA enzyme, thus selecting lymphocytes that have incorporated the vectorcarrying this gene.

10

Inject genetically engineeredlymphocytes with gene forADA enzyme into the patient.

11

Withdraw blood and isolateT lymphocytes from a patient whose DNA lacks the gene for ADA.

8

Allow new virus particles toincorporate the recombinant DNA.

7


Technical difficulties in gene therapy are numerous. Transferring largepieces of DNA into cells is difficult (most genes are large). Inserting agene in a location in the DNA where its protein product will be tran-scribed and translated in a normal way is far more difficult.

The gene therapy described above provides a functional gene that istranscribed and translated by the body cells, producing the missingenzyme in lymphocytes. Because lymphocytes are not the only cells thatneed the ADA enzyme, the patient must also receive injections of theADA enzyme coupled to a molecule that permits it to enter cells. (Thislast step might not be necessary for the treatment of other enzymedefects.) The enzyme controls the symptoms of the disease, but it is not acure because the underlying disease is still present. Gene therapy forADA was first successfully used on a 4-year-old girl in 1990. A secondpatient, a 9-year-old girl, began receiving treatments in 1991. Bothpatients are being closely monitored, and their immune systems are nowworking properly. However, because the genetically engineered cells aremature lymphocytes, which have only a limited lifetime, repeated injec-tions of genetically engineered cells are needed.

To get around this problem, in the hope of bringing about a morelasting cure, some Italian researchers have tried using both geneticallyengineered lymphocytes (as described above) and genetically engineeredbone marrow stem cells. Stem cells divide to form all the developed typesof blood cells (see Chapter 12) and they maintain this ability throughoutlife. Therefore, after repaired lymphocytes die off, stem cells withrepaired DNA could divide to provide new, ADA-functional lymphocytes,possibly for the lifetime of the individual. This type of therapy was begunon a 5-year-old boy in 1992, and since then several other children havereceived this treatment.

Questions of safety and ethics. There are legitimate safety concernswith human gene therapy. For example, any virus used as a vector mustbe capable of entering human cells. Might such a virus cause a disease ofits own? To preclude this possibility, the viruses used in human genetherapy have been from viral strains with genetic defects that renderthem incapable of reproducing and spreading to other cells. Might ran-dom insertion into the host DNA destroy some other gene? Methods arebeing developed for directing the insertion location, but it is still largely arandom event. In 1999, gene therapy clinical trials were halted in theUnited States when an 18-year-old boy died after receiving a viral vectorfor gene therapy for a metabolic disease. The reasons for his death werenot apparent, so clinical studies were halted until issues of safety couldbe addressed. The boy’s father has testified at a U.S. Senate hearing thatthe boy and his family were not fully informed of the dangers of theexperiment. Others have raised ethical objections to the use of the term‘gene therapy’ in clinical trials when most of the experiments that havebeen done so far have not been designed to cure any condition, only toalleviate symptoms (or to test the safety of the procedure itself).

Gene therapy also raises other ethical concerns. New recombinantDNA procedures are very expensive to develop. This raises ethical issuesof fairness: will the benefits of genetic engineering be available only to

102

CONNECTIONSCHAPTER 12

those who can afford them? Should government programs providethem through Medicare and Medicaid? Should insurance cover theiruse? How can society’s health care resources best be distributed? Ifmedical resources are limited, should an expensive procedure used onone person take up needed resources that could cover inexpensive treat-ments of other diseases for many people? These particular questionsare not unique to genetic engineering; they apply to any expensive formof medical treatment.

Genetic engineering may someday become commonplace in humancells. In theory, gene therapy could be practised either on somatic cells oron gametes. If it were performed on somatic cells, the effects of the genetherapy would last as much as a lifetime, but no longer. For example,insertion of the functional allele for insulin into the pancreatic cells ofpatients with diabetes might cure them of the disease, but they wouldstill pass on the defective alleles to their children. A general consensushas been reached that using gene therapy on somatic cells has an ethicalvalue if it is used for the purpose of treating a serious disease.

If successful gene therapy is performed on germ cells, then thegenetic defect will be cured in the future generations derived from thosegerm cells. In addition to all the ethical questions raised earlier, genetherapy on germ cells raises many additional ethical questions. Mostmedical ethicists today advise caution and waiting in the case of germ-cell gene therapy on humans until we have more experience with genetherapy on somatic cells or in other species.


THOUGHT QUESTIONS

1 The use of growth hormone for thetreatment of shortness (not dwarfism) inotherwise healthy children is controversial,but its testing for this purpose wasapproved in 1993 by the Food and DrugAdministration. When does a phenotypiccondition unwanted by its bearer become adisease to be treated? Who decides?Should the use of human growth factorproduced by engineered bacteria toincrease someone’s height be allowed? Is this simply another form of cosmeticsurgery, similar to breast implants or face-lifts?

2 If a person dissatisfied with his or herphenotype suffers from lack of self-esteemon that account, does the lack of self-

esteem justify a procedure to correct thephenotype? (This same argument is raisedto justify traditional forms of cosmeticsurgery.) Do parents have the right toanticipate for a child what the futureeffects on self-esteem will be with andwithout corrective procedures? For aphenotype such as height that developsover a period of years, at what age is itappropriate (if ever) to evaluate thephenotype and decide upon correctivemeasures?

3 A procedure such as gene therapy isexpensive. Who should pay for it? Is genetherapy a limited resource? Does givinggene therapy to one patient thereby depriveanother of medical care?


Molecular Techniques Have Led to NewUses for Genetic Information

Molecular biology is an interdisciplinary field that focuses on DNA.Although there are many other kinds of molecules, molecular biologistsare concerned mostly with DNA. Molecular biology techniques can tellus a lot about human genetics, and several marker systems have nowbeen discovered for studying human DNA. The first of these marker sys-tems, restriction-fragment length polymorphisms, is described here.More recently other markers, with names such as expressed sequencetags, microsatellites, and single-nucleotide polymorphisms, have beendiscovered.

Each person has a unique DNA sequence. If it were practical tosequence a person’s whole genome, his or her DNA could definitivelyidentify a person. The human genome is far too long for it to be usefulfor such identification, but the DNA marker techniques that have been souseful in mapping gene regions have also proved useful in distinguishing,with a high probability, any person from another except for identicaltwins. Two frequent uses of this technique are in the identification of sus-pects in police investigations and in disputes over paternity.

The first DNA marker: restriction-fragment lengthpolymorphismsIn 1980 a new mapping technique was devised that could readily be usedin human studies, as well as in studies on other species. DNA contains, inaddition to genes, noncoding regions that vary in length from one indi-vidual to another. Short sequences of nucleotides, 3–30 bases long, arerepeated over and over anywhere from 20 to 100 times. These are calledshort tandem repeats. Several thousand different such repeats are nowknown in humans, each with a unique sequence not found elsewhere inthe genome. When DNA containing variable numbers of repeats is cutwith a restriction enzyme, fragments of DNA of various lengths are pro-duced (Figure 4.5A). Variations (also called polymorphisms) in thelengths of the fragments produced with restriction enzymes are knownas restriction-fragment length polymorphisms, or RFLPs (pro-nounced “riflips”). The fragments of different lengths are separated by atechnique called electrophoresis (Figure 4.5B). As we saw in Chapter 3(Figure 3.8, p. 73), because DNA carries an electric charge it moves in anelectric field. When a DNA sample that has been cut into fragments isloaded onto a gel and electric current is applied, the fragments move.The gel material retards the movement of the fragments somewhat, andthe larger the fragment, the more its movement is retarded by the gel. Inthe time that the electric current is on, smaller fragments will thereforemove farther than large fragments. Because the nucleotide sequence ofeach short tandem repeat is unique, each can be detected by a specificprobe, a piece of DNA with a sequence complementary to the repeatsequence (Figure 4.5C). Probes are specific and cause only those frag-ments to show up that have sequences complementary to the probesequence.

104


Molecular Techniques Have Led to New Uses for Genetic Information 105

Using DNA markers to identify individualsUsing the same DNA marker techniques that we saw above, geneticistscan compare DNA samples from different persons. The samples are cutwith restriction enzymes. Pieces are separated according to size by elec-trophoresis and then transferred to a paper material. Radioactivelylabeled probes complementary to known DNA sequences are then usedto detect the fragments containing particular variable repeats. Thesefragments appear as bands, with their location indicating the fragmentlength. Several probes can be used at once so that many bands show up,not just one or two as in the example shown in Figure 4.5, in which justone probe was used.

Bands at the same position indicate fragments of the same length insamples being compared. If the band patterns are not the same, then itcan be stated with certainty that two samples did not come from thesame person. In the example from a criminal investigation shown in Fig-ure 4.6, person 1 can be eliminated as a suspect because the band patternfrom the evidence is not the same as that from sample 1. The reverse isnot true, however; band patterns that are the same are not an absoluteguarantee that the samples came from the same individual. What arebeing visualized are chunks of DNA of variable lengths, not the DNA

Figure 4.5Restriction-fragment lengthpolymorphisms (RFLPs).

(A) CUTTING DNA WITH RESTRICTION ENZYMES

The pieces differ in length depending on the number ofrepeats that exist within a piece. In this example, the piecefrom the father is shorter because it has fewer repeatsthan the piece from the mother, which is longer because it has more repeats.

(B) SEPARATION BY ELECTROPHORESIS

The mixture of pieces is placed on a gel and exposed to anelectric field. Because DNA has a negative charge, the pieces move toward the positive electrode. In the time that thecurrent is on, smaller pieces travel farther through the gelthan the larger ones do. None of these pieces is visible yet.

(C) DETECTION WITH A PROBE

None of the pieces can be seen; however, they can be detected with a variable-repeat probe tagged radioactively or chemically (bands shown in color). The probe is a small pieceof DNA with a sequence complementary to the sequence of thatvariable repeat, so the probe will bind to those pieces of DNAcontaining that variable repeat. The probe thus does two things: it identifies pieces with that specific repeat and it indicates whether the sequence is repeated a few times (to give a short DNA piece)or many times (to give a long piece). Other probes will find othersequences that are repeated in other chromosomal locations.

chromosomefrom father

chromosomefrom mother

repeat sequence restriction enzyme cut

DNA fragment not boundby the probe

longer piece frommother’s chromosome

shorter piece fromfather’s chromosome

sample loaded onto gel by pipette

power

source

+

DNA from a pair of chromosomes

direction of travel

Figure 4.6Forensic DNA technology. Inthis example, the evidencesample shows the samepattern of bands as DNAfrom suspect 2. There istherefore a high probabilitythat the DNA in the evidenceis from that suspect. Theperson from whom sample 1was taken can be eliminatedas a suspect.


sequences of the chunks. A score is calculated that indicates how likely itis that a randomly chosen person, other than the one tested, could havethe same band pattern.

The likelihood that another, randomly selected person could have thesame banding pattern is made very small in two ways. First, the DNAprobes selected are those that pick up specific DNA markers that are rarein a given population. Also, several DNA probes are used, one afteranother, to produce a composite banding pattern. The probability thatthe bands produced with just one DNA probe are the same for two peopleis equal to the frequency of that DNA marker in the population. If morethan one DNA probe is used, the probability of both band patterns’matching is equal to the population frequency of the first DNA markermultiplied by the population frequency of the second, and so on for mul-tiple DNA probes and markers.

There are many ways in which the banding pattern can yieldflawed or ambiguous results if samples are not properly processed.In samples from crime scenes, there is often DNA from mixedsources, including DNA from several people and from bacteria orfungi. Protein material in the sample may slow the movement of arestriction fragment in the electrophoresis, making the DNA frag-ment appear as though it were larger than it is. Other chemicals inthe samples, such as the dyes in cloth, can interfere with the restric-tion enzymes cutting the DNA. However, when the tests are doneproperly and with the proper controls, they can be very reliable. Inaddition to linking suspects to material taken from crime scenes, themethods can be used to settle questions of disputed parentage. Themethods can also be used to identify the dead when an intact corpseis not available, as in the aftermath of the terrorist attacks in theUnited States on 11 September, 2001.

Using DNA testing in historical controversies An unusual use of this technique helped shed new light on a histori-cal controversy involving Thomas Jefferson, the third president ofthe United States. DNA markers were used to investigate whetherThomas Jefferson could have been the father of children borne byone of his slaves, Sally Hemings. Two oral traditions exist: descen-dants of Hemings’s sons, Eston Hemings Jefferson and ThomasWoodson, believe that Jefferson was their ancestor, while descen-dants of Jefferson’s sister believe that one of her children, Jefferson’snephew, fathered Sally Hemings’s later children. Researchers com-pared Y chromosomal DNA from descendants of two of Sally Hem-ings’s sons with DNA from descendants of one of Thomas Jefferson’suncles. No Y chromosomal DNA was available from Thomas Jeffer-son’s direct descendants because he had no sons who survived tohave children.

The DNA data show that a set of 19 markers (collectively calledthe haplotype) is shared by all five of the descendants of Jefferson’suncle who were tested and by the descendants of Eston Hemings Jef-ferson. The haplotype is not shared by descendants of Hemings’sother son, Thomas Woodson, or by the descendants of Jefferson’snephew, nor was it found in almost 1900 unrelated men. Thus, Jeffer-son may definitively be ruled out as the father of Thomas Woodson.

106

evidence

isolate andpurify DNA

digest DNA with restriction enzyme

separate DNA fragments byelectrophoresis

transfer fragments to nylon membrane(Southern blotting)

add radioactivelylabeled DNAprobes

wash membrane,expose to X-rayfilm, develop

DNA profiles E = evidenceS1, S2 = samples from two suspects

E S1 S2

samples fromtwo suspects

The Human Genome Project Has Changed Biology 107

In the case of the positive match, however, the evidence supports, butdoes not prove, the idea that Thomas Jefferson could have been EstonHemings Jefferson’s father. As we explained earlier, positive matchesindicate probabilities, not definite identity. The researchers state thatbecause “the frequency of the Jefferson haplotype is less than 0.1%,”their results are “at least 100 times more likely if the president was thefather of Eston Hemings Jefferson than if someone unrelated was thefather.” They also state that they “cannot completely rule out other expla-nations of our findings,” but that “in the absence of historical evidence tosupport such possibilities, we consider them to be unlikely.” Interest-ingly, although the authors are very precise in the text of their article, thetitle, “Jefferson fathered slave’s last child,” overstates their results (E.A.Foster et al. Nature 396: 27, 1998).

The Human Genome Project Has ChangedBiology

The complete genetic material of an entire organism is known as itsgenome. In 1986, scientists proposed a project to make a genetic map, orcatalogue, of a prototypical human, including the chromosomal locationof all human genes and the complete DNA sequence of the genome.Many scientists and physicians think that many medical and other bene-fits could flow from knowing the location and sequence of all the genes.Such knowledge would facilitate locating genes that are associated withdiseases or disease susceptibility. It will also make possible the develop-ment of drugs that are much more specifically tailored to block particu-lar molecules. This effort became known as the Human Genome Project.

The Human Genome Project was funded by the U.S. Congress tobegin work in the fall of 1989, and James Watson, co-discoverer of thedouble-helical structure of DNA, was appointed as the first director. Watson stated his belief that the Human Genome Project would tell uswhat it means to be human.

THOUGHT QUESTIONS

1 Thomas Jefferson had daughters whosurvived to have children. Why was theDNA of their descendants not used in thestudy to determine the paternity of EstonHemings Jefferson and Thomas Woodson?

2 The authors of the Jefferson study statethat they “cannot completely rule out otherexplanations of our findings.” What otherexplanations are biologically possible?

3 Think about the study done on DNA fromdescendants of Jefferson’s family and SallyHemings’s sons. Why is the title of thestudy, “Jefferson fathered slave’s lastchild,” an overstatement of the results?

4 In the study on Jefferson’s descendants,why did the researchers test DNA at 19DNA marker sites, rather than just at oneor two sites?

It should be noted, however, that although we talk of the humangenome sequence, the DNA sequence of each person is unique. There isno one DNA sequence that is representative of every human, just as noone person could be said to represent all humans in any other method ofdescribing people. It is estimated that one person differs from another inabout 0.1% of the 3 billion base pairs in the human genome. Peopleshare the same genes but the nucleotide sequences of those genes vary indifferent alleles.

Sequencing the human genomeOne of the stated goals of the Human Genome Project was to determinethe human DNA sequence. When we read in the newspaper or hear ontelevision about a genome being sequenced, what does this mean? The‘sequence’ of DNA is the order in which the four nucleotide bases (seeChapter 2, p. 56) appear from one end of the DNA molecule to the other.Because DNA is an unbranched molecule, the sequence of bases can be‘read’ from one end to the other.

Determining the order of nucleotides by using fluorescent dyes.Because the amount of DNA in even one chromosome is enormous, it isnot practical to work with the whole length of a chromosome in deter-mining sequences. The maximum size of pieces that can be sequenced iscurrently about 500–700 bases long. The chromosomes are therefore sep-arated and each is cut into overlapping pieces with restriction enzymes.Each piece is inserted into a plasmid which enters a bacterium. The bac-teria then divide repeatedly and make large quantities of one piece at atime, as we saw on p. 98 for bacterial production of human insulin.

The nucleotide sequence of each of the pieces can then be deter-mined using an established method (called the di-deoxy method) basedon DNA synthesis. The DNA is used as a template for synthesis of newDNA strands in a test tube, as outlined in Figure 4.7. The overall result isthe production of a series of smaller pieces, each piece one nucleotidelonger than the next. Each of the small pieces is then separated by elec-trophoresis. The pieces are made visible with a fluorescent dye, a differ-ent color used for each of the four nucleotides. Unlike the specific probesused with DNA markers, fluorescent dyes make all of the pieces visiblethat end in that nucleotide. The sequence of bases in the DNA fragmentcan thus be read from the gel: the base found at the end of the shortestpiece is first (traveled farthest in the gel), followed by the base found atthe end of the next longer piece (traveled the second farthest in the gel),and so forth.

Mistakes can occur in either copying or sequencing, and repeatingthe process does not always give the same answer, so the technique mustbe repeated several times by different laboratories until a consensussequence is established. After the sequence of each piece has been deter-mined, the pieces must be arranged in their original order to get theoverall sequence. Remember: this sequence analysis has been carried outon only one fragment of a chromosome at a time. The next challenge isto piece together the sequenced fragments, which is part of the mappingprocedure discussed below.

The non-coding DNA. Most of the human chromosomal DNA does notcode for genes, however, and the Human Genome Project included the

Chapter 4: Genetic Engineering and Genomics108



sequencing of these non-coding regions. The non-gene DNA consists of‘spacer’ sequences that are never transcribed, and other kinds ofsequences that are transcribed but never translated. The function of mostof these non-gene sequences is currently unknown, and the wisdom ofspending an estimated $15 billion on their sequencing is a question onwhich opinion, even among scientists, differs widely. These non-codingregions, however, have turned out to be the locations of many of the DNAmarkers discussed earlier, which have allowed us to find where specific

Figure 4.7Discovering the nucleotidesequence of a piece of DNA.

1. PRECURSORSG C A T direction of synthesis

C G T A T A C AG T C AGG T C

primer to start synthesis

single-stranded DNA tobe sequenced

normaltriphosphateprecursors(A, T, C, G)

small amountof abnormal precursors(A and T and C and G)

add enzyme

G

A

G

C

C

T

C

AT

A

T A

G

G

GCAT AGCAT ATGTCAGCAT ATGTCAGTCCA

GCAT ATGCAT ATGTGCAT ATGTCAGT

GCAT ATGTCGCAT ATGTCAGTCGCAT ATGTCAGTCC

GCAT ATGGCAT ATGTCAGGCAT ATGTCAGTCCAG

A piece of single-stranded DNA to be sequenced is added to a test tube with an enzyme to activate DNA synthesis and the four precursor triphosphates (black A, T, C and G). Also added are small amounts of chemicals similar to each of the triphosphate precursors, which can add to the growing chain but cannot then bond to the next precursor. Each of the four types of abnormal precursors is labeled with a differently colored fluorescent dye: red As, green Ts, blue Cs and orange Gs.

2. DNA SYNTHESIS

DNA synthesis is then allowed toproceed. When a normal, black precursor is added to the template, the chain keeps growing. When, by random chance, an abnormal precursor gets added instead, synthesis of that chain stops, leaving a strand shorter than the strand being sequenced. Each chain is one nucleotide longer or shorter than the others. Each short sequence endswith a fluorescently tagged molecule.

3. ELECTROPHORESIS

The pieces can then be separated by size using electrophoresis. In the time that the current is on, the fragment that consists of the primer plus a single nucleotide (A in this illustration) will travel the farthest. The fragment that is the primer plus two nucleotides (A + T) will travel not quite as far, and so forth.

4. READING THE SEQUENCE

A fluorescence detector reads each band of the gel, detecting the color of the dye labeling that band.

powersource

+ +

direction in whichDNA moves

duringelectrophoresis

sequence ofnewly

synthesizedDNA

A T G T C A G T C C A G

amo

un

t o

f fl

uo

resc

ent

colo

r d

etec

ted

distance from bottom of theelectrophoresis gel

+


genes are located. Other scientists suggest that these non-coding regionswill also turn out to be important for other reasons. For example, thenon-coding regions are the binding sites for proteins, such as the SRYprotein (see Chapter 2, p. 48), that regulate DNA folding, and thus regu-late when a gene is transcribed.

The human genome draft sequenceIn February 2001 two groups simultaneously announced completion of adraft of the sequence of the human genome. One group, the InternationalHuman Genome Sequencing Consortium, involving laboratories fromthe United States, Britain, Japan, France, Germany, and China, pub-lished their results in Nature (409: 860). The other group, a biotechnol-ogy company called Celera Genomics, published their results on thesame day in Science (291: 1304). The draft covers about 94% of the esti-mated 3 billion bases in the complete genome. Of those 3 billion bases, 1billion have been sequenced to completion, including all of those on thesmallest paired chromosomes, chromosomes 21 and 22. The other 2 bil-lion bases contain gaps and areas where different efforts at sequencinghave resulted in different answers.

Completion of the draft sequence supported some previously estab-lished hypotheses, but also produced some surprises. Some key results are:

1. About 95% of the human genome represents non-coding DNA, alarge proportion of which is composed of repetitive sequences. Lessthan 5% of the human genome is composed of genes, sequencesthat code for RNAs or proteins. It has been known for a while thatthe complexity of an organism does not correlate with the size ofits genome. Much of the excess size is due to these non-coding,repeat sequences. Detailed knowledge of these sequences is open-ing up a new resource for studying evolution. These sequencescan be likened to living fossils carried within each of us. They arealready used in population genetic studies examining the migra-tions of human populations.

2. The actual number of genes is smaller than previously estimated. Inhumans it is difficult to predict which sequences represent genes,for reasons we discuss later. Thus, although the draft sequence ofthe human genome has been published, the number of genesremains unknown. The estimate of the number of genes is cur-rently between 30,500 and 35,500. (Previous estimates had beenbetween 50,000 and 100,000 genes.) The numbers of genes in thefruitfly (Drosophila melanogaster) and the roundworm(Caenorhabditis elegans) have been ascertained; comparisonsreveal that humans are likely to have only twice as many genes aseach of them.

3. The protein products of many human genes remain unknown. Ithas been found that many of the known genes can be translated indifferent ways to produce alternative protein variants from thesame gene (see Figure 4.10, p. 117). Thus, although we have onlytwice as many genes as fruitflies, we may have five times as manydifferent proteins.

110



4. A very high percentage of our genes are not unique to humans butare closely similar to comparable genes from other species. In fact,only 1% of human genes have no sequence similarity to any otherorganism. Our genes are similar to 46% of the genes in yeast,among the simplest organisms whose cells have a nucleus.Changes within genes over time provide clues to rates and pathsof evolution.

5. More than 200 human genes and their protein products have beenfound to have significant similarity to those in bacteria. Thesegenes are not found in intermediate organisms such as fruitflies,and one school of thought suggests that these genes jumped frombacteria to humans or vice versa.

6. Mutation rates differ in different parts of the genome. They are alsohigher in males than in females, although the reason for such adifference is not known.

7. Within each gene, there is an average of 15 sites at which differentindividuals carry a different nucleotide, or at which the same indi-vidual may have a different nucleotide on each chromosome in apair. These variations, called single-nucleotide polymorphisms,are greatly expanding how many alleles we think are possible fordifferent genes. In addition, these small changes may affect thephysiology of the organism possessing them. Some of these poly-morphisms are associated with disease; most are not, but areinstead associated with small changes in protein function or regu-lation. Knowledge of such small-scale variations continues tochallenge our concepts of terms such as ‘heterozygous’, ‘domi-nant’ and ‘recessive’, and ‘allele’. It also makes it clear that there isno such thing as the human genome sequence. The genomesequence within each individual is unique.

In April of 2003, only two years after publication of the draft sequence,the sequence of the human genome was completed. Its publication in thejournal Nature was timed to coincide with the fiftieth anniversary of Wat-son and Crick’s article describing the double helical structure of DNA.

Mapping the human genomeAnother goal of the Human Genome Project was to map the humangenome. Mapping a species’ genome means identifying the chromosomallocation of each gene and the order of the genes relative to one another.Just determining the sequence of a piece of DNA does not tell you itslocation in the genome. The molecular techniques developed as part ofthe Human Genome Project have accelerated the mapping and identifi-cation of genes more generally.

One way to map a large piece of DNA is to cut the same long piecewith two different restriction enzymes, derive the sequence of each of thepieces, then use computers to discover how the two sets of pieces over-lap. Figure 4.8 shows how sequence data from overlapping fragments ofDNA are used to derive the original order of the fragments. Figure 4.8Ashows two sets of fragments of DNA produced by cutting a DNA samplewith different restriction enzymes. The first restriction enzyme cut the


DNA into six pieces only; the second resulted in eight pieces. The bases inthe sequences of each of the eight pieces can be lined up to match thebases in the six pieces. Can you see how you would use this idea to deter-mine the order that the six pieces had originally been in? Now turn thepage and look at Figure 4.8B.

In our example the largest piece contains 40 bases. Actual DNApieces for sequencing are around 500 bases in length. Because the piecesare so much longer and there are so many of them, computers areneeded to line up the overlaps. The accuracy of the method increaseswith the length of the overlapping region. The longer the sequence of theoverlap between two pieces, the higher the probability that the sequencewill appear only once in the genome, allowing the unambiguous assign-ment of the position of the two pieces relative to each other.

Celera used this approach first in 1995 with the complete sequencingof the genome of the bacteria Haemophilus influenzae. The sameapproach was used successfully on the genomes of the 599 viruses, 31eubacteria, and 7 archaebacteria that were sequenced between 1995 and2002. They believe that the same approach will work for mapping thehuman genome.

But there are obstacles to applying this approach to mapping thehuman genome. One obstacle is size; the human genome is about 25times larger than any previously sequenced genome, although it is farfrom being the largest genome known. (One species of single-celledamoeba has a genome 200 times larger than humans!) Another obstacleto accurate reassembly is the fact that much of the non-coding DNA inthe human genome is composed of repeated sequences of nucleotides.This enormously complicates the job of putting pieces into unambiguousorder. Species whose genomes had previously been sequenced do notcontain these repeats, so it was much easier to determine which piece

went where in these genomes. The International Human

Genome Sequencing Consor-tium therefore used DNAmarkers in addition tosequence overlap to map thelocations of the pieces. In thetechnique used by the Con-sortium, the total DNA in thegenome was split into 29,298overlapping large fragmentswith a variety of restrictionenzymes. Each large piecewas further split into piecesof a size that could besequenced. Sequencing of thesmall pieces has been pro-ceeding at the same time asthe mapping of the large frag-ments, and one advantage ofthis approach is that differentlaboratories can be simulta-

112

Figure 4.8(A)Combining the sequencesof small pieces into thesequence of the originalwhole chromosome. Hereare the fragments of asequence cut with twodifferent enzymes. Can youpiece them together toreconstruct the completesequence? Don’t turn thepage until you’ve tried it!

In this example, a DNA sequence of 150 bases is cut with two different restriction enzymes, producing the following fragments, each of which has been sequenced.

Try to piece these fragmentary sequences together and determine the entire sequence of 150 bases, before you turn the page.

Fragments from the first restriction enzyme:

GGTCGGCTATGTAACGAGTTGCC

TCTTGTTCCTAGCTTGTCAACCGGGGATGAATGTTTACTG

CACGCGGACCGTCGGTTCAT

GTCGCAGAGCCTATTGCGAGAAGT

GCCCACCTT

TTATTGAGTTGATGCTCGACGTAGCCAGACTTAA

Fragments from the second restriction enzyme:

ACCGGGGATGAATGTTTACTGGTCGCAGAG

CCTATTGCGAGAAGTGGTCGGCTA

CTTGTCA

TGATGCTCGACGT

CGTCGGTTCAT

AGCCAGACTTAACACGCGGAC

TGTAACGAGTTGCCGCCCACCTTTTATTGAGT

TCTTGTTCCTAG


neously working on different pieces of the puzzle. Indeed, the location ofeach of the large fragments within the genome has now been mappedand the map is publicly available. Mapping of all of the small pieces isstill proceeding.

Because Celera started with all small pieces, the Consortium main-tains that Celera will not be able to reassemble the sequences of theirsmall pieces without referring to the publicly available data posted by theConsortium. Celera maintains that because the Consortium map andsequence data are publicly available, Celera should use it to help assem-ble their small pieces more quickly. Why continue to insist on the slowway, when those data can now be used in a more rapid way?

The Consortium requires rapid, public disclosure of all data. Theirdecision to publish a draft sequence as fast as possible was driven, intheir words, by “concerns about commercial plans to generate propri-etary databases of human sequences that might be subject to undesirablerestrictions on use” (Nature 409: 863). These worries have to do with thestated intentions of Celera Genomics to require others to pay for accessto their databases.

Some ethical and legal issuesMany of the issues already covered in Chapter 3 regarding genetic testingwill become more commonplace as molecular genetics continues tochange medicine. How does an individual’s right to privacy balanceagainst family members’ desire to know the results of genetic tests or aninsurance carrier’s or employer’s desire to cover or to hire only employ-ees who will remain healthy? How does an individual’s desire to controltheir own reproduction balance against possible eugenic aims of societyor against further stigmatization of disabled people? How can geneticcounseling be value-free while providing education about genetics andnot just about the testing procedure itself?

When the Human Genome Project was funded, scientists saw theneed for examination of the ethical, legal, and social issues (anticipatedand unanticipated) that would be raised by the research. One percent ofthe funding was set aside for this effort. The issues just mentioned areamong those being studied, but there are many others. Social workers,anthropologists, ecologists, ethicists, and others are working together toexamine the issues raised by the study of genetic variation in humanpopulations and by the integration of genetic information into healthcare as well as into non-clinical settings. Others are studying the ways inwhich socioeconomic factors, race, and ethnicity influence people’sunderstanding, interpretation, and use of genetic information. Simulta-neously, new genetic information continues to change our concepts ofrace and ethnicity (see Chapter 7). Others are examining how geneticknowledge and concepts interact with different philosophical and theo-logical traditions. Many of the working groups have composed reportswith their answers to many of these questions and their guidelines forthe use of genetic information. These reports are available at the Web sitewww.genome.gov.

In addition, the data derived from the Human Genome Project raisequestions of ownership and patent rights. Who owns the human genome

CONNECTIONSCHAPTERS 3, 7


or the sequence of any particular gene? If a researcher localizes a gene toa particular chromosome, can that researcher patent the information?Can a gene sequence be copyrighted in the manner of a book? Can thegenes themselves be patented? Certain biotechnology companies stand toprofit greatly from the marketing of gene sequences, tests for genesequences, or cures for various genetic diseases, but the sharing of infor-mation on gene sequences seems at first glance to threaten their compet-itive position. Several corporations intend to determine as many genesequences as possible and then copyright them and sell the informationat a profit. Other scientists think that the human genome should be pub-lic information, and that scientists should share this information cooper-atively, particularly if public money in the form of research grants hasbeen used in production of the knowledge. A middle ground is develop-ing, wherein most sequences are posted in data banks with public access,but fees are sometimes charged for that access.

114

Figure 4.8(B)Here is the completesequence of 150 bases.Geneticists usually workwith hundreds of fragmentsat once, each of them longerthan this entire sequence, sothe task of piecing themtogether is much moredifficult.

When the two sets of fragments are lined up in this way, the order of the bases in the first row is the same as the order of the bases in the second row.

and the complete sequence is therefore as follows:

TCTTGTTCCTAGCTTGTCAACCGGGGATGAATGTTTACTGGTCGCAGAGC-



CTATTGCGAGAAGTGGTCGGCTATGTAACGAGTTGCCGCCCACCTTTTAT-



TGAGTTGATGCTCGACGTAGCCAGACTTAACACGCGGACCGTCGGTTCAT



deduced sequencefragments from first enzymefragments from second enzyme

THOUGHT QUESTIONS

1 To what extent do you agree with Watson’sstatement that sequencing the humangenome will tell us what it means to behuman? Suppose you knew the exact genesequence of part or all of your genome;what would you really know aboutyourself?

2 If only stretches of DNA 500–700 baseslong can be sequenced at a time, howmany of these small sections of DNA mustbe sequenced to determine the sequence ofthe entire human genome? (Think alsoabout the overlaps required to piece thesequences together; assume an average of10% overlap.)

3 Will the DNA sequence of the humangenome tell us what traits are controlledby each part of the sequence? Will it tell uswhich sequences represent genes andwhich sequences represent spacers?

4 If you have a certain rare geneticcondition, and scientists use cell samplesfrom your body to determine the gene’sDNA sequence, what rights (if any) doesthis give you to the information? Do thescientists have the right to publish yourgene sequence, or any part of it? Is it aninvasion of your privacy? Can thescientists sell the information? If they do,are you entitled to a share of the profits?

Genomics Is a New Field of Biology Developed as a Result of the Human Genome Project 115

Genomics Is a New Field of BiologyDeveloped as a Result of the HumanGenome Project

The Human Genome Project also funded the sequencing of the genomesof many other species. This may seem odd at first because the name ofthe project specifies the human genome, but there were several reasonsfor including these other species. The study of the genomes of specieshas become an entire new area of biology called genomics. This field hasarisen to help unfold the mysteries of human genes now that thesequences and mapping are nearing completion. One focus of genomicsis the identification of individual human genes. The combination ofmolecular biology and computer science that has been necessary to navi-gate through the tremendous amounts of data produced by the variousgenome projects is called bioinformatics.

BioinformaticsJust as the NASA space program led to many unexpected ‘spin-off’ tech-nologies in the 1960s and 1970s, the Human Genome Project is doing soas well, with new computer technologies and genetic engineering havingwide applications outside genetics. DNA sequencing and mapping wouldnot have been practical before the advent of large computers. Althoughthe techniques for determining sequences of short pieces of DNA arerather simple (see Figure 4.7), finding the overlaps that indicate how thesmall sequenced pieces were originally arranged (see Figure 4.8) requiresmassive computer power. Then, when the longer sequences have beendetermined, storing the data has necessitated the development of largerand larger computer databases and new methods for searching them.Genomics requires the development of new types of computer software.The need for people who are trained in both molecular biology and com-puter science who can work with these data has made bioinformatics afast-growing new field of employment.

One research project within bioinformatics has been the develop-ment of computer programs to locate genes within a genome. In the past,as we have seen, scientists worked from a trait, back to finding a gene.Now that the genome sequence is complete for many species and nearlycomplete for humans, the method of gene discovery has changed. Nowpeople are examining the sequence data itself and trying to determinewhich parts may be genes, without any prior knowledge of a trait or afunction for those genes. Many such genes have already been found inbacteria and yeast, and they are referred to as “orphan genes” because, atthe time of their discovery, no function was known. (The later identifica-tion of their function is part of the research program of functionalgenomics, described later.)

Within bioinformatics, people are programming computers to scanthe sequence data to locate genes, meaning areas that code for RNAs andproteins. To do so, programmers must discern ‘rules’ of the genetic code:what characteristics of a sequence distinguish a coding region from anon-coding region? The computer search for genes within sequences iscalled gene scanning.


Gene scanning in different organisms. Interestingly, most genes startwith the codon ATG and end with one of three ‘stop codons’: TAA, TAG,or TGA. If the nucleotides A, T, G and C were distributed randomly, eachof the stop codon triplets would be expected to occur on average every 43

or 64 bases. But nucleotides are not distributed randomly within genes;they are retained in a non-random pattern as a result of evolutionbecause they code for a product conferring advantage to the organism. Inbacteria, genes are typically 300–500 codons long, are contiguous, and donot overlap. In addition, bacteria have very little non-coding DNA. Thesefactors make gene scanning in bacteria relatively easy. A computer canscan the sequences that follow any ATG and find those areas where thenext stop codon occurs a few hundred bases further along.

Gene scanning is much more difficult in other organisms, namely thenucleated organisms (eucaryotic organisms; see Chapter 5). In contrastto bacterial species, they have long, non-coding stretches of nucleotides(called introns) dispersed among much shorter regions that correspondto codons. While the coding regions (called exons) are roughly the samelengths in different species, the size of the non-coding introns is muchgreater in humans than in other species (Figure 4.9).

Within the human genome, less than 5% is located within genes; fur-thermore, within these human genes, only about 5% of the nucleotidescomprise coding sequences. This makes it difficult to use raw sequencedata to predict which nucleotide regions represent genes. Thus, gene-scanning programs are continuing to be refined to include up-to-dateknowledge about the characteristics of the ‘departures from randomness’within genes in species other than bacteria. One such departure fromrandomness is called ‘codon bias:’ not all codons are used equally by agiven species. For example, the amino acid alanine can be coded by fourdifferent codons in humans; furthermore, three of those are used muchmore frequently than the fourth. In a non-coding region of the genome,all of the codons have an equal probability of being represented, but in acoding region the one codon is present less frequently.

The presence of non-coding regions within genes is clearly a com-plication for gene scanning. Of what benefit could it be to an organismfor its genes to be interrupted by such non-coding regions? It is thesenon-coding stretches that have allowed the shorter coding segments torecombine to form new genes. This provides a mechanism for rapidgenetic change (more rapid than by mutation). New genes are pro-

duced by the novel assembly ofparts. There is another way inwhich the division of genes intomany coding regions is adap-tive, and that is in providing amechanism by which slightlydifferent versions of a proteincan be made in different tis-sues, adapted to the cellularenvironment and function ofthat tissue. An example isshown in Figure 4.10. The

116


Figure 4.9Single strands of DNAshowing the differences ingene structures in bacteriacompared with eucaryoticcells. (A) Bacterial genescontain only coding regions;that is, the DNA is alltranscribed to mRNA. (B) In eucaryotic cells non-coding regions that arenot transcribed are locatedwithin the coding regions ofgenes. (C) In humans (a eucaryotic species) theamount of non-coding DNAis much greater than theamount of DNA that codesfor a protein product.

DNA

DNA

DNA

gene (1000 nucleotides)

gene

coding region

coding regions(exons)

noncoding regions(introns)

coding regions noncoding regions

human Factor VIII gene (200,000 nucleotides)

(A) one strand of bacterial DNA

(B) one strand of DNA from a nucleated cell

(C) one strand of human DNA


human gene for a protein called a-tropomyosin contains many codingregions scattered among non-coding regions. This gene can be tran-scribed to mRNA in different ways, so that in the cells of one tissue oneset of coding regions is used, and in the cells of another tissue, a differentset of coding regions is transcribed. This results in different mRNAs in thedifferent cells, and therefore in slightly different proteins after translation.Each protein is still a-tropomyosin, but with a slightly different aminoacid sequence and therefore a slightly different functional capability.

Although scientists think these large non-coding regions withingenes are adaptive for the organism, they do present a significant obsta-cle to identifying genes by gene scanning. In fact, it does not appear that gene-scanning programs alone will be able to identify all of thegenes in a eucaryoticgenome. Hence, theHuman Genome Pro-ject also funded workon the genomes ofother species, so thathuman genes could belocated by comparisonwith the genes of otherspecies, a field nowknown as comparativegenomics.

Comparative genomicsWhen scientists compare sequences of genes from one species with thosefrom another, they are working in the field of comparative genomics. Thesize of the genome of many species has been determined. As we saw ear-lier, the overall size does not always correlate with the complexity of theorganism. This is due to the very great differences in the amount of non-coding DNA in various genomes, so that overall size does not correlatewith the numbers of genes present.

As we have just discussed, genes are much easier to identify in somespecies than in others. Once a gene has been identified and its sequencedetermined in one species, there is often enough sequence similarity forits counterpart gene (or genes) to be located in other species. This is themajor reason why other species’ genomes were also examined as part ofthe Human Genome Project. Another reason was that sequencing thegenomes of other species allowed scientists to develop the technologythat was later used to analyze human genome sequences.

Many human genes have been located by their similarity to yeastgenes. A yeast cell, like a human cell, has a nucleus and many of its geneshave remained very similar to the counterpart genes in humans. Animalsare even more similar and one animal that is proving to be quite useful incomparative genomics is the pufferfish, Fugu rubripes. Its genome is onlyone-seventh the size of the human genome, yet it is estimated to have thesame number of genes. Because of its small genome size, gene location ismuch easier in pufferfish, and may subsequently allow mapping of thehuman counterpart genes. The mouse genome is almost complete, and

Figure 4.10Within human genes all ofthe nucleotides aretranscribed into RNA butonly some of the RNAnucleotides are translatedinto protein.

mRNA for tropomyosin in smooth muscle

mRNA for tropomyosin in fibroblast cells

mRNA for tropomyosin in brain cells

mRNA for tropomyosin in striated muscle

DNA strand

a-tropomyosin gene

codingregions

non-codingregions

transcription and splicing

many more human genes will be found by comparison with those in themouse. Many known genes in mice are located in the same order on theirchromosomes as they are on human chromosomes, and this correspon-dence is extremely helpful in mapping genes. The mouse genome, how-ever, is even larger than the human genome, so the problems of workingwith a large genome still pertain. See our Web site for information on thesizes of the genomes of various species (under Resources: Genome sizes).

Aside from its usefulness in locating human genes, comparativegenomics has produced new data for evolutionary biologists. Speciesthat have a common ancestor have more genes and more nucleotidesequences in common than species that do not. Unfortunately, the scien-tists working on a particular species have often independently devisedthe database for each species’ genome. Consequently, another goal ofbioinformatics is to devise ways of making the different databases com-patible and interactive, thereby facilitating comparative genomics.

In addition to finding similarities between species, comparativegenomics has led to the realization that within a species there are groupsof genes that share large portions of their sequences. These ‘gene fami-lies’ are presumed to have evolved from a common ancestral gene. Find-ing one gene in the family enables the others to be located, and mostoften the different protein products of the family members to be identi-fied. For example, the human hormones oxytocin and vasopressin (bothproteins) belong to the same gene family, and they have very similaramino acid sequences and genes that code for them. The same is true ofthe oxygen-carrying proteins hemoglobin and myoglobin.

Functional genomicsIn Chapter 3 we described how Archibald Garrod and other scientistsstudied “inborn errors of metabolism,” disease conditions caused bychanges in biochemical pathways. The study of similar changes in bacte-ria or yeast have often led to the discovery of entire chains of biochemicalreactions. In the past, scientists looking for the molecules involved insuch a biochemical pathway, would start with a trait and work backwardsto a protein. Pedigrees such as we saw in Chapter 3 would be linked withdifferent forms of a protein. After purifying the protein and discoveringits amino acid sequence, its gene sequence could be inferred. Genesequencing turns this whole process around. Genes are found by linkageto DNA markers, and only later is the protein product found. However,finding a gene, mapping its location, sequencing it, and even deriving theamino acid sequence of its protein product, will not tell us its function.

New sequences can be compared with those whose function isalready known. This is the field of functional genomics. Species that canbe easily manipulated experimentally have been most useful in discover-ing gene functions. The zebrafish is a vertebrate that reproduces rapidly,and many of its internal structures are visible in the living fish becauseoverlying structures are transparent. For these reasons, zebrafish havebecome an experimental species of great interest to scientists workingon the genetics of development. An even simpler species, the yeast Sac-charomyces cerevisiae, has been found to share many genes withhumans. Gene functions that were discovered in yeast have proved to



www

have parallels with disease-associated gene mutations in humans. Forsome examples see our Web site, under Resources: Yeast genes. The func-tions of the yeast genes are discovered by several methods. One is toexamine which genes are transcribed to mRNA when the yeast under-goes a particular response or function; another is to inactivate (mutate) agene and see what effect this has. Once the sequence of a gene is known,it is relatively easy to mutate it by manipulating the DNA causing achange in the protein product, which is now not functional. The oppositeapproach can also be used: extra copies of the gene can be inserted andobservations made of the change in function under different environ-mental conditions. These approaches are not confined to yeast, but arealso done to discover gene functions in mice and other species.

Earlier in this chapter we saw how a gene could be added to agenome, using a vector. In Figure 4.4 we saw how the functional gene forADA was added to the genome in human cells. This method adds a gene,but in an unpredictable location within the genome. The non-functionalgene is still present, and indeed one of the possible dangers of the tech-nique is that the new gene may get added in a place that disrupts someother gene. More recently, methods have been developed for changingthe sequence of a specific gene. In theory, this technique could be used torepair a non-functional gene, to mutate a gene in a specific way, or to dis-rupt a functional gene. A vector is used to carry into the cell a piece ofDNA partly complementary to the gene to be altered. The introduceddouble-stranded DNA becomes substituted for the gene region as a resultof crossing-over at two sites where the insert and the gene have the samesequences (Figure 4.11). If the inserted DNA is non-functional, as shownin this example, the normal gene is disrupted. The effect of deletion ofthat gene’s protein can then be studied in the offspring cells (yeast or tis-sue culture of human cells). If the gene disruption is carried out on a cellfrom a very early stage of development, an entire organism can developthat is lacking the gene and its protein product. (This topic, part of stemcell research, is covered in more detail in Chapter 12.) Mutated mice withparticular genes nonfunctional or ‘knocked out’, or mice with overex-pressed genes produced by the insertion of additional copies of a func-tional gene, have led to important clues to the functions of human genes.

Families of genes have been found within species that have struc-turally related protein products but very often quite different functions.This has led scientists to realize that gene duplication and mutation canoccur first, and that new functions can follow. Of course, a gene that ispresent in just a single copy cannot change to a new form (possibly witha new function) without giving up its original form and function.A duplicated gene, however, can undergo changes in one copy(possibly evolving new functions) while the other copy remainsunchanged.

ProteomicsJust as the complete DNA sequence of an organism is its genome,the complete protein content of that organism is its proteome. Proteomics is the study of how the protein content changes overtime in a cell and in an organism, how it differs in different tissues,


Figure 4.11In human genes, differentcombinations of codingregions can be transcribed toproduce different mRNAs indifferent tissues.

recombinationat two sites

chromosomal DNA

disrupted gene

vector DNA

www



and how it relates to the health and function of the organism. Proteinsare synthesized as a result of transcription of genes and translation ofmRNA, as we saw in Chapter 3. However, there are further modificationsto a protein after it has been translated that affect both its activity and itsconcentration. No protein stays in a cell forever; all are degraded andremoved. We will see more about these aspects of protein function incells in Chapter 12.

Knowing the DNA sequence of genes has hastened the discovery ofthe amino acid sequence of proteins. Computer programs use the knownenergies and bond angles of chemical bonds to turn amino acid sequencedata into molecular models of the three-dimensional shape of a proteinor portion of a protein. Having the ability to visualize these shapes bycomputer graphics has led to new strategies for the design of medicines.In the past, natural products and synthetic compounds were randomlytested in functional assays to see which would work for a particularneed. Now small molecules can be designed to exactly fit a critical enzy-matic site of a protein. Once the molecule has been designed by com-puter simulation, medicinal chemists then synthesize it and biologiststest to see whether it has the desired outcome of blocking the protein’sfunction. The action of such drugs is far more specific, and the drug willtherefore have fewer side-effects than traditionally developed drugs, forreasons we will study in Chapter 14.

To synthesize a protein with even a slightly different structure can bevery difficult and costly. However, once the sequence for the gene codingfor that protein is known, it becomes relatively easy to modify the pro-tein by changing the sequence of its gene. Roughly the same technique asthat in Figure 4.11 is used, but the inserted piece of DNA differs from thenormal piece by only a few nucleotides. Such modifications can, forexample, lead to the development of proteins that are stable under awider variety of conditions. These proteins find a variety of industrialapplications. Stain removers in laundry detergents, altered enzymes forfood processing, and cleanup of pollution, are just a few examples.

Rather than studying one protein change at a time, proteomics alsohas another goal: to study all of a cell’s proteins in the aggregate. Such agoal has been unattainable in the past, and is a big factor explaining whyreductionism (reducing a problem to its simplest form) has been awidespread experimental approach in biology. Proteomics is in itsinfancy, but promises to be a much more integrative approach.

120

CONNECTIONSCHAPTERS 3, 12, 14

THOUGHT QUESTIONS

1 In what ways are humans poor subjects forgenetic research? In what ways arehumans good subjects? Which of yourreasons are purely biological, and whichhave ethical components?

2 Why are certain traits studied in somespecies and not others?

3 Will genomics allow the findings in onespecies to be applied in other species? Whyor why not?

Summary to Chapter 4 121

Concluding Remarks

As genomics has discovered genes with useful properties within onespecies, genetic engineering has given us the tools to transfer those genesinto another species. The Human Genome Project has discovered thattransfer of genes from one species to another does also occur in nature.Viral and bacterial genes are found in the human genome, for example.Because we have almost always studied the effect of one gene or one pro-tein at a time, transferring a gene into a new genomic environment maylead to different results from those that we expect. As we develop thetools to alter genomes, proteomics may give us the ways to study theeffects of such changes throughout the cell. We also need to be mindful ofeffects at the level of the whole organism and effects of genetic engineer-ing on ecosystems as well, which we will explore further in Chapter 11.

Chapter Summary

• Restriction enzymes cut DNA into fragments with known sequences attheir ends. Restriction enzymes that produce fragments with single-stranded sticky ends are used in genetic engineering to splice new genesinto genomes.

• Variations in the lengths of these fragments are called restriction-fragmentlength polymorphisms or RFLPs. RFLPs have helped in finding the loca-tion of many genes, as well as in the identification of individuals and ingenetic engineering.

• Genetic engineering consists of inserting functional genes into cells,thereby altering the cell’s genotype. The recipient cells may be bacterialcells that may then acquire the ability to make certain human proteins, orthey may be human cells that acquire a functional allele and are injectedinto a patient as gene therapy.

• Bacterial plasmids are used to carry genes into a new species.

• A genome is the total genetic information carried by a particular organ-ism. The Human Genome Project has now produced a draft sequence andmap of the human genome.

• DNA markers of various kinds have allowed the mapping of genes tolocations within the genome. Markers also allow the identification ofindividuals.

• Genomics is the study of the genome, either the comparison of genomesof different species or as a method of discovering gene functions.

• Bioinformatics combines computer science and molecular biology in theanalysis of genomes and the identification of genes within a genome.

• Proteomics is the study of all of the proteins present within a cell.



CONNECTIONS TO OTHER CHAPTERS

Chapter 1 Genetic engineering and gene therapies raise many ethical issues.

Chapter 2 The genome is the blueprint for the proteome.

Chapter 3 We have learned a lot about human genetics by studying comparativegenomics.

Chapter 5 Comparative genomics is a new tool for discovering the evolutionaryrelationships among organisms.

Chapter 6 Comparative genomics may give us important data for use in studyingclassifications.

Chapter 7 The amount of possible variation within each gene is much greater thanwas previously thought.

Chapter 11 Genetic engineering of crop species is increasing agricultural productivity.

Chapter 17 The same DNA testing techniques as those that are used to identifyindividual humans can also identify the bacterial species involved in newinfectious outbreaks and can sometimes also identify its source.

Chapter 18 Comparative genomics is increasing our knowledge of biodiversity.

1. If one individual human differs from another in0.1% of the genome, how many bases are different?

2. In the following stretch of DNA, how manyfragments will result from digestion with theHaeIII restriction enzyme shown in Figure 4.1?How many will result from digestion with EcoRI?

strand 1

A T C C G T A G G C C T A A C C A T C C T A G T G C

T A G G C A T C C G G A T T G G T A G G A T C A C G

strand 2

3. Why are restriction enzymes that producefragments with ‘sticky ends’ more useful in geneticengineering than restriction enzymes that producefragments with ‘blunt ends?’

4. Could the following sequence be used as an insertinto genomic DNA? Why or why not?

strand 1

A A G C T T A A C G G A T T A G C A A G C

C G A A T T G C C T A A T C G T T C G A A

strand 2

5. Could the following sequence be used as an insertinto genomic DNA? Why or why not?

strand 1

A A G C U U A A C G G A U U A G C A A G C

C G A A U U G C C U A A U C G U U C G A A

strand 2

6. When a plasmid is being cut with a restrictionenzyme in preparation for inserting a DNAfragment, the plasmid needs to be cut with thesame restriction enzyme as was used to make theDNA fragment. Why?

7. Can DNA marker band patterns be used to identifymaternity, as well as paternity?

8. Can DNA marker testing be used to identifyindividual organisms in other species besideshumans?

PRACTICE QUESTIONS

genetic engineering and genomics - garland · pdf file · 2017-07-14genetic...

Documents